267 results about "Adverse weather" patented technology

The invention discloses a multi-degraded image restoration method based on semantic guidance. According to the method, for degraded images captured by a vehicle-mounted camera under severe weather conditions, potential features of the images are extracted by adopting a trunk network based on Transform, multi-modal semantic information is extracted in combination with a CLIP visual language model, and semantic guidance is provided for different degradation types through a dynamic text prompt generation mechanism. A self-adaptive feature fusion module is designed, channel attention and space attention mechanisms are combined to realize effective integration of multi-modal features, and a degradation feature extraction and fusion module is introduced to enhance the generalization ability of the model. According to the semantic guidance multi-type image recovery network SGIRN provided by the invention, the global modeling capability of the Transform and the cross-modal representation capability of the CLIP visual language model are combined, so that high-quality recovery of various weather degradation types is realized.

The invention discloses a data processing method and system for intelligent road detection and related equipment, and relates to the technical field of traffic engineering and artificial intelligence crossing. In the road detection process, multi-sensor fusion equipment with dynamic adjustment is adopted to collect multi-source data; the real-time environment perception model based on deep reinforcement learning dynamically optimizes the parameters of the multi-modal sensor, and improves the data acquisition quality in severe weather and environment. Millimeter-level alignment of multi-source data is realized by adopting a spatio-temporal model, and a generative adversarial network is introduced to enhance the multi-source data after spatio-temporal alignment so as to supplement missing data details and improve data quality. Disease intelligent identification is carried out on multi-source data through a sample learning model and a meta-learning algorithm, the sample learning model solves the problem of insufficient labeling samples of micro-disease data, the meta-learning algorithm is quickly generalized on the basis of a small amount of labeling data, micro-diseases such as early cracks and hidden road structure layer defects are identified, and the accuracy of identification is improved. And the accuracy of road disease identification is improved.

The invention discloses an intelligent scheduling method and system for road construction, and relates to the technical field of intelligent scheduling, and the method comprises the following steps: analyzing to obtain a road construction knowledge spectrogram and first scheduling data, and compensating the first scheduling data; according to the method, the historical construction data is analyzed, the road construction knowledge graph is constructed, the construction process, resource investment and construction period arrangement can be dynamically adjusted according to the real-time change of weather by introducing the real-time meteorological data for the complex association among scattered experience, rules, cases and entities, and the dynamic adjustment is realized through meteorological compensation, so that the construction efficiency is improved. Non-productive consumption such as mechanical idling and staff waiting caused by bad weather can be reduced, experience reuse and scientific decision are realized by constructing a road construction knowledge graph, environmental changes are coped with through dynamic compensation of meteorological data, and intelligence, refinement and high efficiency of construction scheduling are finally realized through the synergistic effect of the two.

The invention relates to the technical field of intelligent traffic systems, in particular to a traffic volume prediction method under adverse weather conditions, which comprises the following steps: acquiring and preprocessing multi-source heterogeneous data of roads in a target area; the multi-source heterogeneous data comprises meteorological data, historical traffic flow data, real-time traffic sensor data, dynamic road event data and road network topology data; performing space-time alignment on the preprocessed multi-source heterogeneous data, and determining a fusion feature matrix; based on a pre-trained CNN-LSTM-Transformer hybrid prediction model, inputting the fusion feature matrix, and outputting a traffic volume prediction result in a specified time period in the future; and generating a dynamic traffic control instruction set according to the traffic volume prediction result. The problems that in the prior art, the data dimension is single, spatial-temporal feature extraction is insufficient, and prediction control is disjointed are solved.

The invention belongs to the technical field of high-voltage line grounding equipment, and particularly relates to a lockable grounding wire clamp device mounted by an unmanned aerial vehicle and a method for introducing backflow in a high-voltage line into the ground, and the lockable grounding wire clamp device mounted by the unmanned aerial vehicle can be carried by the unmanned aerial vehicle to complete grounding operation of the high-voltage line in maintenance. The lockable grounding wire clamp device is time-saving and safe for on-site constructors, a unique locking and unlocking mechanism composed of a lock cylinder, a clamping jaw assembly, a spring bolt driving piece and an unmanned aerial vehicle mounting assembly enables a high-voltage wire to enter a high-voltage wire claw embracing area, and then high-voltage wire claw embracing can be completed under the self-gravity effect of the lockable grounding wire clamp device; according to the invention, connection and locking of the high-voltage line are realized, the locking mode is relatively reliable, unlocking can be realized only by pulling up the unmanned aerial vehicle mounting assembly, and safe grounding in adverse weather such as strong wind is realized. When the unmanned aerial vehicle needs to be taken down, unlocking can be achieved by connecting the unmanned aerial vehicle with the unmanned aerial vehicle mounting assembly and pulling the unmanned aerial vehicle mounting assembly upwards, and unlocking is convenient and fast.

The invention relates to a method for cross-view geographic positioning of an unmanned aerial vehicle in severe weather. According to the method, a ConvNeXt-based twin network framework is provided for the matching challenge of an unmanned aerial vehicle and a satellite image under a severe weather condition, so that the positioning capability of the network under complex weather is improved. By introducing the weather simulation module, the network can adapt to view angle changes in severe weather such as rain, fog, snow and the like. According to the method, a local mode network and a direction attention module are utilized, and environment information and multi-direction context information are combined, so that the extraction capability of the network on characteristics under cross-view and different weather conditions is enhanced. Besides, in order to improve the robustness of the features under different view angles, a multi-branch feature fusion strategy is adopted, and fused feature vectors are spliced through a classifier. And finally, rapid matching of the unmanned aerial vehicle image and the satellite image is realized by using the constructed geotag database, and the positioning precision in GNSS denial and severe weather environments is effectively improved. Compared with the prior art, the method has higher robustness and accuracy in a complex environment.

A vehicle including a windshield, a flexible display screen, a flexible display screen attachment component, at least one processor, and at least one memory configured to store instructions is disclosed. The instructions, when executed by the at least one processor, cause the at least one processor to (i) generate a set of images from sensor data received from one or more sensors mounted on a body of the vehicle; and (ii) display the set of images on the flexible display screen attached to the flexible display screen attachment component while covering the windshield. The set of images has effects of adverse weather conditions or poor lighting conditions removed.

The invention relates to the technical field of watering control, in particular to an intelligent watering control method for alfalfa seed production, which comprises the following steps: collecting humidity data of current soil through a soil humidity sensor, analyzing a soil humidity state and generating a soil humidity result; evaluating the required water quantity according to the soil humidity result, and adjusting the output quantity of a water pump to obtain an adjusted watering plan. The current soil humidity state of the farmland is obtained by monitoring the soil humidity in real time, so that the output quantity of the water pump is dynamically adjusted, intelligent regulation and control of the watering plan are achieved, the irrigation accuracy is improved, and waste and insufficiency of water resources in the watering process are avoided; meanwhile, the watering plan is optimized by means of air temperature and rainfall data, so that the irrigation action is closely combined with the actual weather condition, the irrigation timeliness is improved, and the negative effects caused by adverse weather conditions are reduced.

The invention discloses an image restoration method and system coping with various severe weather environment factors, and relates to the technical field of image restoration, and the method comprises the steps: recognizing an image weather type, and carrying out the test time self-adaption of a model; inputting an image into the adaptive model to obtain a feature vector of the image, and decoding the feature vector through a decoder; and transmitting the image to a soft reconstruction layer to obtain a restored image. According to the invention, a supervised learning mode is adopted to train a universal network capable of coping with multiple weathers. A soft prompt excitation model is introduced in the fine tuning stage to enhance the generalization ability to cope with different weathers. Explicit and implicit interaction is utilized to enhance the prompt effect, hidden information of prompts is explored through low-rank decomposition, and by adding contrast loss, prompts approach to similar tasks and leave away from opposite tasks, and the feature coding capacity is improved. According to the method, the influence of various severe weathers on the image can be effectively processed, the image quality is improved, and high-quality input is provided for subsequent defect identification and auxiliary analysis.

The invention relates to the technical field of information processing, in particular to an anti-interference fusion sensing system for a multi-source heterogeneous sensor, and aims to solve the problem that the sensing performance of the multi-source heterogeneous sensor is reduced in severe weather such as rain and fog. The system comprises a data acquisition and synchronization module, a multi-modal feature extraction and weather interference suppression module, an adaptive multi-modal fusion module, a target detection and tracking module and a sensing result output module. According to the method, robust features are extracted from interfered laser radar and camera data through resistance robust feature learning, and millimeter wave radar information is combined; dynamically adjusting the weight of each modal according to weather through a self-adaptive fusion module; the sensing precision and reliability of the vehicle in complex environments such as rain and fog are effectively improved, and driving safety is guaranteed.

Examples of the present disclosure provide a computer-implemented system, comprising instructions for performing operations including: estimating locations in a region of a map where features associated with water, snow or ice are likely to be formed under certain adverse weather conditions; simulating one of the adverse weather conditions; generating the features associated with the simulated one of the adverse weather conditions at the locations; determining a response of a perception stack of an autonomous vehicle (AV) to the adverse weather conditions observed at the locations; determining a reaction of the AV to the features generated in the region, the reaction of the AV being a function of a configuration of the AV; in response to the reaction, updating the configuration; repeating the determining the reaction and the updating the configuration until a desired reaction is obtained; and exporting a final configuration corresponding to the desired reaction to a physical AV.

The invention discloses a video recognition-based man-machine cooperative processing method for ditches and rivers in an urban area, and relates to the field of video recognition or understanding, and the method comprises the steps: constructing a data collection and analysis closed loop covering the whole periods before, during and after heavy rainfall: collecting images before and after treatment through an unmanned aerial vehicle and a camera before rainfall, generating a treatment operation influence map for quantitatively evaluating the treatment effect; in rainfall, the fixed camera is used for continuously monitoring the water level and the flow velocity; after rainfall, images are collected again, and the abnormal river reach is recognized. According to the method, camera data is taken as a boundary condition, a hydrological model is driven to perform inversion in combination with a governance influence map, missing heavy rainfall process data is deduced, and finally, accurate attribution of abnormal reasons is realized, and a differentiated governance scheme is generated. According to the method, the problem of monitoring data missing in severe weather is effectively solved, and quantitative evaluation of the treatment effect and scientific diagnosis of the abnormal cause of the river channel are realized.

The invention relates to the technical field of expressway early warning, and discloses an expressway multi-source risk factor dynamic fusion early warning system under severe weather, which comprises a multi-source risk factor acquisition module, a patrol vehicle and a mobile monitoring subunit integrated with a vehicle-mounted terminal through an along-line fixed monitoring subunit, collecting parameter data of severe weather, road conditions and traffic flow; the data preprocessing module adopts a machine learning algorithm to eliminate abnormal values, complement missing values and standardize the missing values to generate standardized risk data; the machine learning dynamic fusion module constructs a three-layer evaluation index system, a fusion weight is generated through an improved AHP model, nonlinear fusion is realized through a fuzzy integral fusion model, and a comprehensive risk value is output; and the early warning grade judgment module generates an early warning grade through threshold comparison, and finally the early warning grade is converted into standardized early warning information containing early warning types, influence road sections and duration by the early warning output module. The method provides powerful guarantee for road traffic safety.

The invention discloses a LiDAR-4D radar 3D target detection method based on dynamic denoising and dual-domain gating fusion, and the method achieves the robust 3D target detection in severe weather through the three steps: dynamic denoising, adaptive fusion, and precise detection. According to the method, a semantic guidance mechanism is introduced into a feature extraction layer, the perception ability of the network to foreground target points and the inhibition ability of the network to background noise points are enhanced, and meanwhile, the denoising threshold value can be dynamically adjusted by integrating the real-time point cloud density and weather category information through a denoising strategy of the denoising threshold value self-adaptive to environment change, so that the real-time point cloud density and the weather category information can be dynamically adjusted. Therefore, the balance between noise suppression and foreground point reservation is realized under different complex weather conditions, the fusion weight of the laser radar and 4D radar modality can be adaptively adjusted explicitly according to the real-time weather condition through a double-domain gating mechanism fusing external weather perception information, the perception ability of feature fusion to the external environment is enhanced, and the real-time weather perception information is obtained. And the rationality of the fusion strategy and the overall robustness are improved.

The application relates to the technical field of information processing, in particular to a multi-source heterogeneous sensor anti-interference fusion perception system, which aims to solve the problem of decreased perception performance of multi-source heterogeneous sensors in adverse weather such as rain and fog; the system comprises a data acquisition and synchronization module, a multi-modal feature extraction and weather interference suppression module, an adaptive multi-modal fusion module, a target detection and tracking module and a perception result output module; the system extracts robust features from the interfered laser radar and camera data in combination with millimeter wave radar information through adversarial robust feature learning; and then the adaptive fusion module is used to dynamically adjust the weights of various modes according to the weather; the application effectively improves the perception accuracy and reliability of vehicles in complex environments such as rain and fog, and guarantees driving safety.

Described herein relates to a solar panel mounting apparatus and methods thereof for optimizing solar energy capture and / or protecting solar panels from adverse weather conditions. The apparatus may include a rotation drive module that adjusts the orientation of solar panels throughout the day based on time-of-day data to maximize sunlight exposure. The apparatus may also comprise a plurality of integrated sensors to monitor sunlight intensity and / or environmental conditions, providing real-time input to a control unit that may continuously automate panel positioning. Additionally, the apparatus may comprise a safety drive module that allows the solar panels to collapse into a protected position within the mounting structure during extreme weather events including high winds or hail. The retraction may be triggered automatically by a weather detection module or manually, via override controls. These operations may ensure both optimal energy production and / or enhanced panel durability, making the system suitable for varied environmental conditions.

The present invention relates to an erosion control apparatus and methods of using and installing the apparatus. The apparatus is constructed to prevent erosion of soil during typical weather or tidal conditions and adverse weather events. The apparatus can include an anchored marsh installation operative to stabilize a tidal marsh and shoreline.

The invention relates to the technical field of computer vision and image processing, in particular to a container detection method and system oriented to cross-modal adaptive fusion. The method comprises the following steps: acquiring a visible light image, an infrared thermal imaging image and millimeter wave radar point cloud data in a port scene; constructing a multi-modal adaptive fusion deep learning model containing an environmental perception gating network; the environment sensing gating network analyzes the current environment parameters in real time, and dynamically calculates and outputs the fusion weight of each mode; based on the fusion weight, adaptively fusing the features extracted from each modal; and synchronously identifying the container number and detecting the physical state, the safety state and the operation state of the container by using the fused features. According to the method, the accuracy and robustness of port container identification and state detection under severe weather conditions such as fog, rain, snow and night are remarkably improved, and the safety and efficiency of port automatic operation are guaranteed.

The invention belongs to the technical field of traffic target detection, and discloses a traffic target detection method, system and model architecture under severe weather conditions. The method comprises the following steps: preprocessing an input traffic scene image; performing multi-scale feature extraction on the preprocessed image, and obtaining multi-scale feature representation through AMSConv operation; carrying out bidirectional fusion on the extracted multi-scale features to obtain fusion features; the fusion features are processed, and category prediction, bounding box coordinates and confidence of the target are generated; and optimizing bounding box regression based on InfoAware-IoU measurement, and dynamically weighting a high information sample. Through multi-scale feature extraction, efficient detection head design and information perception regression optimization, the problem of visual degradation under severe weather conditions is effectively solved, and the accuracy and real-time performance of traffic target detection are improved.

We provide an imaging lens system, camera module, in-vehicle system, and mobile device with sufficient field of view and brightness for sensing inside and outside a vehicle at night or in adverse weather conditions. [Solution] The imaging lens system 11 consists of a front group Gf, an aperture diaphragm STOP, and a rear group Gr, comprising a first lens L1 with negative power whose image side faces concave toward the image, a second lens L2 with object side faces concave toward the object, a third lens L3 with positive power, a fourth lens L4 with power, a fifth lens L5 with power, a sixth lens L6 with power, and a seventh lens L7 with power. Among the lenses with positive power from the first lens L1 to the seventh lens L7, the third lens L3 has the greatest power, and when the focal length of the first lens L1 is defined as f1 and the focal length of the entire optical system is defined as f, the following condition (1) is satisfied. -2.5

This invention discloses a multi-sensor fusion longitudinal control method and device for unmanned IGV vehicles. The method involves setting a pre-parking position within a predetermined distance from the target bridge equipment. The unmanned vehicle acquires the target parking distance from its current position. Then, it uses multiple sensors from the vehicle's onboard motor speed sensor and IMU inertial navigation system to fit the traveled distance and the distance to be braked, summing them and comparing the sum with the target parking distance. If the sum is less than the target parking distance, PID speed control is applied; otherwise, emergency braking control is initiated. This invention solves the problem of inaccurate parking of unmanned transport vehicles due to poor weather and road conditions, achieving precise parking with a positioning accuracy of 5cm even in adverse weather and road conditions.

This invention discloses a vehicle-road cooperative information interaction method and system based on intelligent transportation, belonging to the field of vehicle-road cooperative technology. The method includes S1: multi-source data acquisition, S2: 3D environment modeling, S3: structured analysis of traffic conditions, S4: multi-source collaborative fusion, S5: vehicle-end collaboration, and S6: adaptive optimization configuration. This invention establishes a multi-source clock synchronization benchmark, accurately measures and compensates for network transmission delay, eliminates time deviations caused by link asymmetry, and achieves motion distortion correction of laser point clouds and spatiotemporal alignment of multi-sensor data. This effectively eliminates motion distortion of lidar in dynamic traffic scenarios, enabling high-density point clouds to accurately reflect the instantaneous state of the road, balancing perception accuracy and real-time processing efficiency, improving perception reliability under adverse weather conditions, and achieving lane-level precise cooperative control and high-precision vehicle positioning capabilities.

This utility model discloses a low-temperature resistant and interference-resistant vehicle lamp, including a supporting shell. An isolation cover is fixedly installed on the side of the supporting shell, and a mounting frame is fixedly connected to the outer side of the supporting shell. A fixing screw is inserted into the side of the mounting frame, and an inner isolation plate is fixedly snapped onto the inner side of the supporting shell. Advanced UV electroplating technology concentrates the light pattern, greatly improving luminous efficiency. This not only reduces glare affecting other road users but also increases the illumination distance, allowing drivers to see the road ahead more clearly at night or in adverse weather conditions, effectively ensuring driving safety. The cover is made of UV-resistant material, increasing wear resistance and effectively preventing scratches. During long-term use, even under various harsh environments and external friction, it maintains good appearance and light transmission, extending the lamp's lifespan and reducing the frequency and cost of lamp replacement.

The utility model provides a photovoltaic inverter support, which comprises a supporting foundation, a bottom layer frame and a mounting frame, and is characterized in that the bottom layer frame is erected on the supporting foundation and is relatively fixed with the supporting foundation; wherein the installation frame is arranged on the side, away from the supporting foundation, of the bottom layer frame, the installation frame is connected with the bottom layer frame through a first supporting beam and a second supporting beam, the first supporting beam and the second supporting beam intersect, and the included angle between the first supporting beam and the second supporting beam ranges from 45 degrees to 60 degrees. In the application, the mounting frame is connected with the bottom layer frame through the first support beam and the second support beam which are arranged at an angle, so that the weight of the photovoltaic inverter and possible external forces such as wind load and snow load can be borne, and deformation or toppling under various severe weather conditions can be avoided.

A vehicle includes a traction control system. The vehicle includes a controller configured to indicate a recommended speed on a display based on precipitation deposits, temperature, and speed limits, and to activate the traction control system before the vehicle intersects the vector. The indication and activation can be in response to an expected route of the vehicle intersecting an expected precipitation deposit of a weather prediction vector received from a source external to the vehicle.

The invention belongs to the technical field of power grid fault analysis and early warning, and particularly relates to a power grid fault risk early warning method based on a fault quantitative relation model, a storage medium and electronic equipment. According to the scheme, a fault quantitative relation model of dynamic and non-linear correlation between the fault rates of various power grid equipment lines and the meteorological levels of various severe meteorological types is established through preferential function fitting, and meteorological level thresholds corresponding to various power grid equipment in various severe meteorological types are scientifically set; clear and quantitative decision criteria are provided for starting fault defense measures or not, power grid fault risk early warning analysis is carried out, and early risk early warning of power grid equipment line faults in a severe meteorological environment is accurately and quickly realized. And key data support can be provided for response measures such as preparation of a power grid fault defense plan in advance, calling of emergency standby resources such as energy storage, configuration scale and the like.

The invention discloses an all-weather unstructured orchard visual navigation system, which is characterized in that an improved CycleGAN image processing module is introduced to perform enhancement processing on an image acquired under a severe weather condition, interference factors such as rain, snow and fog in the image are effectively removed, the image quality is improved, and an improved YOLOv11-seg network is combined, so that the visual navigation precision of the orchard is improved. And high-precision and high-stability segmentation of the unstructured road region is realized. Meanwhile, a continuous, smooth and controllable navigation path is constructed based on a path line generation method of pixel mutation detection and cubic B-spline fitting, and path planning precision and operation stability in an orchard scene are guaranteed. The constructed all-weather unstructured orchard visual navigation system comprises three modules: an all-weather adaptation module based on the improved CycleGAN, a road segmentation module based on the improved YOLOv11-seg, and a navigation path line generation module.

The utility model discloses a honeysuckle planting supporting device, and relates to the technical field of agricultural planting. The honeysuckle flower planting frame comprises a rectangular mounting frame, and honeysuckle flowers are planted in the mounting frame; the fixing frames are fixedly arranged at the top of the mounting frame at equal intervals, and the fixing frames are in an isosceles trapezoid shape; the supporting rods are detachably arranged on the two bevel edges and the top of the fixing frame, and the supporting rods are used for supporting honeysuckle branches; the connecting rod is arranged at the end, away from the fixing frame, of the supporting rod. The main branches of honeysuckle are supported, the branches are effectively supported through the design of the supporting rods and the connecting rods, the overall photosynthesis efficiency and the flowering and fruiting rate are improved, and therefore the yield and the quality are improved, the supporting rods are detachably designed, installation and maintenance are convenient, and under the severe weather condition, the supporting rods can be used for supporting the main branches of honeysuckle. A plastic film can be conveniently laid to protect honeysuckle plants, and the practicability and adaptability of the device are enhanced.

A computer-implemented method includes using an image captured by a vehicle camera on the vehicle to determine a vehicle visibility distance for the vehicle with a computing device on the vehicle. The determined vehicle visibility distance is compared to a predetermined desired visibility distance and a first drone is used at the vehicle visibility distance away from the vehicle when the vehicle visibility distance is less than the predetermined desired visibility distance. An image captured by the vehicle camera is combined with a first drone video image captured by the first drone to provide a combined image having a total visibility distance of a sum of the vehicle visibility distance and a first drone visibility distance.